Better maternal outcomes with testing for placental growth factor when pre-eclampsia is suspected

Reference

Duhig KE, Myers J, Seed PT, et al, for the PARROT trial group. Placental growth factor testing to assess women with suspected pre-eclampsia: a multicentre, pragmatic, stepped-wedge cluster-randomised controlled trial. Lancet 2019:393(10183):1807-1818.

Synopsis

This was a complex trial carried out in 11 UK maternity care centers. Women with a singleton fetus at 20 0/7 to 36 6/7 weeks' gestation (N = 1019) who were suspected of having pre-eclampsia were included. Women were identified at both routine prenatal visits and visits for acute symptoms. Suspected pre-eclampsia was defined as new-onset hypertension (or worsening of existing hypertension), dipstick proteinuria, right upper quadrant pain, headache with visual disturbance, fetal growth restriction, or abnormal maternal blood tests (thrombocytopenia, liver function, kidney function). Women were excluded if pre-eclampsia was already diagnosed at presentation. After a concurrent baseline period of 6 weeks for all centers, maternity units were randomized, representing clusters. All enrolled women had testing for placental growth factor (PlGF), whether the results were revealed to the clinicians or not, and the intervention was to reveal the results to the maternity providers for integration into the national guidelines for management of pre-eclampsia. Prior research has shown a negative predictive value of PIGF of 98% (95% CI 93% - 99.5%). Every 6 weeks a cluster was added to the active intervention, such that each maternity unit (cluster) had a before and after period, of variable duration. Results remained concealed in the control condition (before implementation). Women known to have PlGF of less than 100 pg/mL were targeted for increased surveillance. The intervention and control group sizes were unequal due to the cluster randomization, with 573 in the intervention group and 444 in the control group. The primary outcome was time from suspicion of pre-eclampsia to diagnosis, which was significantly shorter in the intervention group (1.9 days vs 4.1 days; time ratio 0.36; 0.15-0.87; P = .027). More important, the composite of (many) severe maternal adverse outcomes was reduced in the intervention group (4% vs 5%; adjusted odds ratio 0.32; 0.11 - 0.96; P = .43). Outcomes in the composite included, for example, eclampsia, severe pre-eclampsia, and placental abruption. Four women, all in the concealed group and with low PlGF, had eclampsia, stroke, and/or cardiac arrest (0 in the intervention group). Blood transfusions were also more frequent in the concealed group. There were no differences in type of delivery or in the many newborn outcomes studied.